Abstract
Septic arthritis of the knee is an uncommon complication of arthroscopic anterior cruciate ligament reconstruction, with a reported incidence ranging from 0.14% to 1.7% in recent publications. In this study, we assess the clinical presentation, management and early outcome of patients with septic arthritis of the knee following anterior cruciate ligament reconstruction. Literature on this uncommon complication is sparse.
The North Shore and Auckland City Hospital Orthopaedic databases were searched and 13 patients were identified as having been treated for septic arthritis of the knee following anterior cruciate ligament reconstruction in the period from July 2002 to August 2006. Their clinical records were reviewed to compile information regarding their presentation and management. Five of these patients were also recalled for clinical follow-up at an average time of 16 months. We reviewed knee range of motion, stability, functional testing in vertical and horizontal jumps and radiographic changes. Clinical outcomes were further assessed using the Tenger, Lysholm and International Knee Documentation Committee Scores.
The patients reviewed had a mean age of 26 years and presented to hospital at an average of 16 days after their autologous ACL reconstruction surgery. All had initial elevation of inflammatory markers with a mean CRP of 189mg/L (68 – 295) and mean ESR of 71mm in one hr (10 – 112.) Mean peripheral WCC on presentation was 12.3 (9.5 – 22.4.) Initial knee aspirates were performed on all patients and yielded a mean specimen WCC of 60,900 x 106/L. Of the 13 patients, six had S. epidermidis, three had S. aureus, two Propioniobacterium acnes and one Serratia marcescens. No organism was cultured from one patient’s aspirate. The study patients underwent an average of two surgical interventions, the first being arthroscopic washout in each case. Six patients subsequently underwent open knee joint washouts, four of these having their cruciate grafts removed.
Of the five patients recalled for clinical review, three rated their IKDC knee performance as being significantly worse than their uninjured side. Mean IKDC scores were 63.5 for the affected knee and 97.3 for the contralateral knee. Mean Lysholm knee score was 71.8 at follow-up. Tenger scores prior to ACL reconstruction averaged 4.4, compared to 5.6 on review. Radiographs demonstrated evidence of arthritis that was not apparent pre-operatively in four of the five review patients. These individuals lacked an average 2.8 degrees of extension and 13.4 degrees flexion in comparison to their contralateral knee. Two patients demonstrated clinical instability on examination. The mean single-legged hopping distance was 62.9% horizontally and 96.4% vertically, when comparing the affected knee to the contralateral side.
Staphylococcus epidermidis was the most common pathogen identified in this study. Most patients presenting with this complication will require two or more operative procedures and a prolonged course of intravenous antibiotics. The symptomatic and functional outcomes of septic arthritis associated with recent ACL reconstruction are highly variable, but were found to be worse in those patients requiring graft removal to eradicate their infections. Despite their young age, most of those patients undergoing clinical review had radiographic evidence of early osteoarthritis.
Correspondence should be addressed to Associate Professor N. Susan Stott at Orthopaedic Department, Starship Children’s Hospital, Private Bag 92024, Auckland, New Zealand
